Llama Guard 3 8B pricing & specs
Meta · released 2024-07-23 · ● verified 2026-10-11
input / 1M$0.484via Cloudflare Workers AI
output / 1M$0.03via Cloudflare Workers AI
context131.1K
max output131.1K
// price history ($/1M tokens)
// cheaper or similar alternatives
| model | in | out | ctx | vs Llama Guard 3 8B |
|---|---|---|---|---|
| Llama 3.2 1B Instruct | $0.02 | $0.02 | 131K | −94.6% |
| Gemma 3 4B IT | $0.04 | $0.08 | 131.1K | −86.5% |
| Gemma 4 E2B IT | $0.04 | $0.08 | 131.1K | −86.5% |
| Qwen3.7 Flash | $0.03 | $0.13 | 1M | −85.2% |
| AutoGLM Phone 9B Multilingual | $0.035 | $0.138 | 65.5K | −83.6% |
// price by host
| host | in $/1M | out $/1M |
|---|---|---|
| Cloudflare Workers AI | $0.484 | $0.03 |
// faq
How much does Llama Guard 3 8B cost?
Llama Guard 3 8B costs $0.484 per million input tokens and $0.03 per million output tokens (cheapest tracked host: Cloudflare Workers AI). At 50M input and 10M output tokens per month that is about $24.50.
What is the context window of Llama Guard 3 8B?
131.1K tokens, with up to 131.1K output tokens per request.
What are cheaper alternatives to Llama Guard 3 8B?
Llama 3.2 1B Instruct ($0.02 in / $0.02 out), Gemma 3 4B IT ($0.04 in / $0.08 out), Gemma 4 E2B IT ($0.04 in / $0.08 out) offer the same input types and capabilities for less.
Is Llama Guard 3 8B open weights?
Yes. It is available from 1 tracked hosts; the cheapest is Cloudflare Workers AI at $0.484 input and $0.03 output per million tokens.
// spec sheet
- input / 1M
- $0.484
- output / 1M
- $0.03
- blended (3:1)
- $0.3705
- context
- 131.1K
- max output
- 131.1K
- input types
- text
- reasoning
- no
- tool calling
- no
- structured output
- no
- open weights
- yes
- knowledge cutoff
- 2023-12
- price source
- Cloudflare Workers AI
// cost calculator
| model | in $/1M | out $/1M | = monthly |
|---|---|---|---|
| Llama 3.2 1B Instruct | $0.02 | $0.02 | $1.20 |
| Gemma 3 4B IT | $0.04 | $0.08 | $2.80 |
| Gemma 4 E2B IT | $0.04 | $0.08 | $2.80 |
| Llama Guard 3 8B | $0.484 | $0.03 | $24.50 |
Base-tier list prices. Long-context tiers, caching and batch discounts are not included.